How AI, automation, robotics and other innovations in ultrasound are improving stroke care

8988

Robert Hamilton (Los Angeles, USA) outlines several key milestones in the history of stroke care, and discusses how recent technological advances in this space will likely benefit physicians and their patients in the years to come.

Throughout the world, technology is playing an enormous role in sculpting the future and improving the health and wellbeing of individuals. Technological advancements, such as artificial intelligence (AI), robotics, and automation, and are being used every day in neurology departments—proving especially helpful in identifying patients at risk for stroke. The COVID-19 pandemic has also had a heavy hand in accelerating the use of digital health applications, robotics, virtual care and wearables among physicians and patients alike.

Historical developments in stroke research

The concept of stroke was first mentioned in the fifth century BC by Hippocrates. At the time, he referred to the signs of paralysis as apoplexy, a Greek term that stands for ‘struck down by violence’, and described the disease as ‘pain [that] suddenly seizes the head in a healthy person, and he at once becomes speechless’.

Centuries later, when human dissections were regularly conducted, knowledge of the anatomy of the brain—and how it changes because of certain diseases and conditions—was swiftly gained. Many medical interventions came about and evolved due to the efforts of Swiss physician scholars Felix Platter and Jakob Wepfer. Platter conducted a brain autopsy after the death of one of his stroke patients and found a ‘phlegmatic humor obstructing the inner passes of the brain’. Later, Wepfer realised that apoplexy was a cerebrovascular disorder and precisely described the main cerebral arteries known as the Circle of Willis.

Fast-forward to the 20th century, and the wealth of research and knowledge around stroke had grown exponentially. Diagnostic technology, such as computed tomography (CT) and magnetic resonance imaging (MRI) scans came about in the 1970s, helping researchers and clinicians better understand brain function. The detailed images of blood vessels and arteries in the brain enabled clinicians to distinguish whether a patient experienced neurological issues like ischaemic stroke or cerebral haemorrhage.

In 1982, transcranial Doppler (TCD) ultrasound was first described in the Journal of Neurosurgery as a safe, non-invasive, real-time way to measure cerebral haemodynamics. A TCD exam is a painless ultrasound method that uses high-frequency soundwaves transmitted through the tissues of the brain to evaluate and measure the rate and direction of blood flow in the Circle of Willis. Since it was initially described, TCD has established itself as an essential tool in several areas of patient care, including detection of cerebral embolus, cerebral stenosis, right-to-left shunt, brain death, and much more.

While the TCD exam can be an important tool for monitoring brain circulation, it is rarely used, because it requires significant training and expertise. A trained healthcare professional must skilfully find the transtemporal window—a thin area of bone that allows the ultrasound waves to penetrate the skull—locate the proper vessels, and then proficiently interpret the signal. TCD expertise is scarce and, without complementary technology to aid in the exam, its use is limited.

The promise of AI, robotics and automation

Technology powered by AI and machine learning are complementing and guiding medical professionals to make more accurate evaluations of patients at risk for stroke. The use of these technologies for stroke care accelerated during the pandemic and autonomous, robotic-assisted ultrasound systems allowed for distance between patients and sonographers, helping to reduce the risk of potential transmission. Additionally, ‘telestroke’ allowed for the acute care of cerebrovascular disease during quarantine.

As examples, a team of researchers in Korea developed a wearable patch that monitors a patient’s blood pressure to continuously measure heart activities in real time. Smartphone apps, such as the Stroke Riskometer, emerged as new tools to assess an individual’s risk of having a stroke in the next five or 10 years, with tips on how to reduce the risk too. None of these advancements would have been possible without the digital health revolution.

As technologies continue to advance, so too will clinicians’ abilities to perform even more sophisticated neurological procedures. Clinical trial data presented at the 2022 American Heart Association’s (AHA) International Stroke Conference (ISC 2022; 9–11 February, New Orleans, USA) assessed the use of robotics- and AI-assisted TCD compared to transthoracic echocardiogram (TTE) for right-to-left shunt (RLS) diagnosis. The multicentre, prospective, single-arm BUBL clinical trial (NovaSignal) evaluated 129 adults who presented neurological symptoms for embolic stroke or transient ischaemic attack (TIA). The AI-driven robotics element allowed for each ultrasound probe to independently scan the temporal area, find the temporal window, and monitor bilateral cerebrovascular blood flow. Results revealed that robotics-assisted TCD was three times more likely to detect RLS in patients with presumed embolic strokes than TTE. Additionally, the autonomous robotic TCD used in the trial was operated by healthcare providers without TCD expertise—confirming the importance of this modern technology integration.

Robotics have been an established and integral part of the healthcare industry for many years and will continue to help enhance patient care in the future. Stroke care has come a long way from Hippocrates to Wepfer to modern day. It continues to evolve with new technologies designed to further enhance physician ability, and aid in more accurate monitoring and diagnosis of patients at risk for stroke and other conditions. Technology will also become a vital element in expanding access to care and helping improve outcomes for even more patients. The continued incorporation of AI, robotics and automation into stroke care pathways has the potential to significantly improve stroke care for generations to come.

 

Robert Hamilton is the chief scientific officer and co-founder of NovaSignal. He is an accomplished entrepreneur, engineer and clinical researcher with a passion for innovative technologies that allow for increased access to care. A biomedical engineer by training, Hamilton is an expert in image/signal processing and machine learning, with extensive experience in cerebral blood flow, traumatic brain injury, stroke and other neurological disorders. He has also achieved more than 100 citations of his work in peer-reviewed publications and conferences, and holds over 50 patent assets related to the core technology developed during his PhD studies.

 

DISCLOSURES: The author disclosed that he is the chief scientific officer and co-founder of NovaSignal—a company that attempts to uniquely capture blood flow data in real time by applying robotics and artificial intelligence to cerebral ultrasound scans.


LEAVE A REPLY

Please enter your comment!
Please enter your name here